henhaobio
Sign in

ScienceGeneticsPeer-reviewed journal article

Analysis of shared heritability in common disorders of the brain

Verneri AnttilaiD, Brendan K Bulik-SullivaniD, Hilary K. FinucaneiD, Raymond Kenney WaltersiD, José BrásiD, Laramie E. DuncaniD, Valentina Escott‐PriceiD, Guido J. FalconeiD, Padhraig GormleyiD, Rainer MalikiD, Nikolaos A. PatsopoulosiD, Stephan RipkeiD, Zhi WeiiD, Dongmei YuiD, Phil Hyoun LeeiD, Patrick TurleyiD, Benjamin Grenier‐BoleyiD, Vincent Antoine ChourakiiD, Yoichiro KamataniiD, Claudine BerriD, Luc LetenneuriD, Didier HannequiniD, Philippe AmouyeliD, Anne BolandiD, Jean‐François DeleuzeiD, , Nicholas A. FurlotteiD

PDFDOIPreprintData
In-depth full-text digest available · Jump to digest ↓

Abstract

Disorders of the brain exhibit considerable epidemiological comorbidity and frequently share symptoms, provoking debate about the extent of their etiologic overlap. We quantified the genetic sharing of 25 brain disorders based on summary statistics from genome-wide association studies of 215,683 patients and 657,164 controls, and their relationship to 17 phenotypes from 1,191,588 individuals. Psychiatric disorders show substantial sharing of common variant risk, while neurological disorders appear more distinct from one another. We observe limited evidence of sharing between neurological and psychiatric disorders, but do identify robust sharing between disorders and several cognitive measures, as well as disorders and personality types. We also performed extensive simulations to explore how power, diagnostic misclassification and phenotypic heterogeneity affect genetic correlations. These results highlight the importance of common genetic variation as a source of risk for brain disorders and the value of heritability-based methods in understanding their etiology.

In-depth summary

Claude Opus 5.5 · max effort · 8 Nov 2026 · from the full text of version 2 (6 Sept 2017) · CC BY 4.0

Genetic correlations across 25 brain disorders (215,683 cases) reveal broad psychiatric common-variant sharing, largely distinct neurological disorders, and migraine as the only significant cross-category link.

Brain disorders are often comorbid and share symptoms, and earlier GWAS found strong genetic correlations among disorders such as schizophrenia, bipolar disorder and major depressive disorder (MDD), but whether the neurology-psychiatry divide reflects distinct common-variant etiology was untested at scale. Pooling uniformly processed summary statistics for 25 disorders and 17 traits, plus power and misclassification simulations, enables the first comprehensive test.

Key results

  1. ShownPsychiatric disorders share much common-variant risk: schizophrenia, bipolar disorder, anxiety disorders, MDD and ADHD form a tightly intercorrelated group, whereas autism spectrum disorder (ASD), Tourette syndrome (TS) and post-traumatic stress disorder (PTSD) look more distinct.

    Mean genetic correlation (rg) = 0.40 among schizophrenia, bipolar disorder, anxiety disorders, MDD and ADHD; schizophrenia significant with most psychiatric disorders; MDD positive (not always significant) with every other disorder tested; anorexia nervosa, OCD and schizophrenia mutually significant; TS correlated with OCD and MDD; PTSD had no significant rg. · Heritability and correlations among brain…; Discussion

  2. ShownNeurological disorders are more genetically distinct and share little with psychiatric disorders; migraine is the exception, correlating with ADHD, TS and MDD, while Alzheimer's, Parkinson's, generalized epilepsy and multiple sclerosis correlate little with other brain disorders.

    Mean cross-category rg = 0.03; migraine-ADHD rg = 0.26 (p=8.81 × 10^−8), migraine-TS rg = 0.19 (p=1.80 × 10^−5), migraine-MDD rg = 0.32 (p=1.42 × 10^−22); migraine-bipolar rg = −0.03 (p=0.406); focal epilepsy had the highest neurological mean rg (0.46, excluding other epilepsy sets) but no significant pair; focal vs generalized epilepsy rg = 0.16 (p = 0.388). · Heritability and correlations among brain…; Discussion

  3. ShownBrain disorders share genetics with cognition and personality: ADHD, anxiety disorders, MDD and TS correlate negatively with cognitive measures, anorexia nervosa, ASD, bipolar disorder and OCD positively, and neuroticism's most significant correlation is with MDD.

    Neuroticism-MDD rg = 0.737 (p=5.04 × 10^−96); schizophrenia negative with intelligence but positive with years of education; Alzheimer's, epilepsy, intracerebral hemorrhage, ischemic and early-onset stroke and migraine significantly negative with cognition; never/ever smoking with MDD rg = 0.33 (p=3.10 × 10^−11) and ADHD rg = 0.37 (p=3.15 × 10^−6). · Heritability and correlations among brain…; Discussion

  4. SuggestedSimulations on UK Biobank genotypes indicate diagnostic misclassification is an implausible sole explanation for the observed correlations, although heavy misclassification depresses observed heritability and thus power to detect sharing.

    120,267 genotyped UK Biobank individuals were permuted to simulate phenotypes; absent any true genetic correlation, misclassification up to 79% would be needed to reproduce observed rg (prior work put the level needed for high rg at 15-30%); modest misclassification or heritability shifts changed rg little. · Study design; Discussion

  5. ShownComparison traits and heritability partitioning add etiological context: coronary artery disease correlates with MDD and stroke, Crohn's disease with no study phenotype, and immune-cell enrichment appears for multiple sclerosis but not Alzheimer's or Parkinson's.

    Coronary artery disease rg = 0.19 with MDD (p=8.71 × 10^−5), 0.69 with ischemic stroke (p=2.47 × 10^−6) and 0.86 with early-onset stroke (p=2.26 × 10^−5); central nervous system (CNS) enrichment for generalized epilepsy, MDD, TS and neuroticism; conserved-region enrichment for bipolar disorder, MDD, schizophrenia and migraine. · Heritability and correlations among brain…; Discussion

Methods, data and reproducibility

Uniformly quality-controlled GWAS summary statistics (European-only meta-analyses where needed) were analyzed with linkage disequilibrium score regression (LDSC) for heritability (liability scale for binary traits) and pairwise genetic correlations, plus stratified LDSC tissue/functional enrichment and UK Biobank genotype-permutation simulations.

  • Brainstorm GWAS meta-analysis summary statistics for 25 brain disorders, with epilepsy, migraine and ischemic stroke subtypes analyzed as nested subsets · human · n = 215,683 cases; 657,164 controls
  • Psychiatric GWAS (PGC, ANGST, TSAIGC): ADHD, anorexia nervosa, anxiety disorders, ASD, bipolar disorder, MDD, OCD, PTSD, schizophrenia, Tourette syndrome · human
  • Neurological GWAS (IGAP, ILAE, ISGC, IHGC, IMSGC, IPDGC): Alzheimer's, epilepsy, intracerebral hemorrhage, ischemic stroke, migraine, multiple sclerosis, Parkinson's · human
  • Behavioral-cognitive GWAS (SSGAC, GPC, CTG, TAG): educational attainment, childhood cognition, IQ, Big Five personality, depressive symptoms, well-being, smoking · human · n = 744,486 (13 phenotypes)
  • Additional comparison phenotypes: BMI and height (GIANT), Crohn's disease (IIBDGC), coronary artery disease (Cardiogram) · human · n = 447,102 (four phenotypes)
  • UK Biobank genotyped individuals permuted to simulate phenotypes under power, misclassification and heterogeneity scenarios · human (UK Biobank) · n = 120,267
  • Stratified LDSC genome annotations from Finucane et al.: top-level tissue-type and functional partitions · human · n = ten tissue-type and 53 functional partitions
Methods & tools
LD score regression (LDSC), Stratified LD score regression (heritability partitioning), Block jackknife standard errors, Liability-scale heritability transformation, Weighted least-squares regression (study-makeup effects), Bonferroni multiple-testing correction, Genotype-permutation phenotype simulation (UK Biobank)
Reproducibility gaps
No analysis or simulation code linked; PGC bipolar disorder and MDD summary statistics marked 'soon'; simulation details deferred to Supplementary Text; other download pages (IGAP, IMSGC, IPDGC, GPC, CTG, IIBDGC, Cardiogram) also listed.
Limitations (6)
  • Modest power of the ASD, PTSD, Tourette syndrome and focal epilepsy meta-analyses limits conclusions about their sharing patterns; focal epilepsy's broad, non-significant correlations may also reflect heterogeneity or diagnostic misclassification. · stated by the authors
  • Generalizability is constrained by small samples for some neurological disorders and by under-representation of circuit-based disorders (restless legs syndrome, sleep disorders, possibly essential tremor) and of personality disorders in psychiatry. · stated by the authors
  • Three ischemic stroke subtypes and NEO agreeableness lacked sufficient heritability and were excluded, and some pairs had standard errors too large to interpret (e.g., migraine with aura vs ischemic stroke, rg = 0.29, p=0.099). · stated by the authors
  • Genetic correlation cannot distinguish shared risk factors, phenotypic overlap, heritable intermediate traits or unequally weighted shared effects; cognition links may reflect ascertainment bias or behavioral effects on testing; locus-level dissection awaits more loci. · stated by the authors
  • Datasets were restricted to European-ancestry meta-analyses, so whether these sharing patterns hold in other ancestries is untested. · noted in this summary
  • LDSC captures only common-SNP sharing; rare or structural variants, such as the pleiotropic ion-channel mutations noted in the paper's background, are not assessed, so low rg does not exclude shared etiology. · noted in this summary

Read this if you study shared genetic risk across psychiatric and neurological disorders, or need to judge how power and diagnostic misclassification shape LDSC genetic correlations.

Citations
1,924
FWCI
152.37
Field percentile
top 1%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISSmeland '25Rietkerk '25Romero '22Radonjić '21Marees '21Zhong '26
77% semantic matchGenetic & Genomic Medicine
Olav Bjerkehagen Smeland, Gleda Kutrolli, Shahram Bahrami+18 more·Nature Neuroscience·Nov 11, 2025· 23 citations